A method and apparatus are disclosed for performing maintenance operations in a system using address, data, and controls which are transported through the system, allowing for parallel and serial operations to co-exist without the parallel operations being slowed down by the serial ones. It also provides for use of common shifters, engines, and protocols as well as efficient conversion of ECC to parity and parity to ECC as needed in the system. The invention also provides for error detection and isolation, both locally and in the reported status. The invention provides for large maintenance address and data spaces (typically 64 bits address and 64 bits data per address supported).
Legal claims defining the scope of protection, as filed with the USPTO.
1. A method of performing data operations in a computer comprising: receiving in a common queue a stream of data operations, the data operations include control commands, address, and data; identifying fast commands, which are required to execute within a particular amount of time; identifying slow commands, which are allowed a longer time for execution than fast commands; separating fast commands from slow commands by sending the fast commands to a fast engine for processing fast operations using a parallel connected satellite, and the slow commands to a slow engine for processing slow operations using a serial chain.
2. The method according to claim 1 , where fast commands include parallel time-of-day (TOD) operations.
3. A method of performing data operations comprising: receiving in a common queue a stream of data operations, the data operations include control commands, address information, and data information; loading the address information into a data word, loading the data information into another dataword, starting a routing process to separate fast commands from slow commands by sending the fast commands to a fast engine for processing fast operations using a parallel connected satellite, and the slow commands to a slow engine for processing slow operations using a serial chain; and performing a read or a write operation using the address and data provided.
4. The method according to claim 3 , where status is returned.
5. The method according to claim 4 which includes the additional steps of: detecting an error, and indicating such error on a particular bit of the status.
6. The method according to claim 5 which includes the additional step of: aborting the operation in the presence of an error, and retrying the operation again.
7. The method according to claim 3 , where data is returned.
8. The method according to claim 3 where the performing a read or a write operation is accomplished using a parallel, broadcast/mux configuration of satellites.
9. The method according to claim 3 where the performing a read or a write operation is accomplished using a serial loop of satellites.
10. The method according to claim 3 where the performing a read or a write operation includes a masked write operation to set only specific bits.
11. The method according to claim 3 where the performing a read or a write operation includes a masked write operation to reset only specific bits.
12. The method according to claim 3 , wherein the fast commands being processed by the fast engine includes transferring the address information and the data information to a local engine.
13. The method according to claim 3 , further comprising using only parity in response to sending at least one of the fast commands to the fast engine, and the slow commands to the slow engine.
14. A computer having maintenance operations, comprising control commands, address and data storage and transfer logic, and having a common queue for a stream of maintenance operations, some of said commands being commands for fast operations which need to execute within a particular amount of time, and other commands for slow operations which are allowed a different and longer time for execution, a fast engine for processing fast operations using a parallel connected satellite and a slow engine for processing slow operations using a serial chain, and differentiators for separating slow from fast operations in the system on critical operations while allowing for the versatility of serial operations for accessibility.
15. The computer according to claim 14 wherein said parallel and serial operations co-exist without the parallel operations being slowed down by the serial operations implemented for accessibility.
16. The computer according to claim 14 wherein is provided a plurality of engines for fast and slow operations.
17. The computer according to claim 16 wherein said engines utilize a common shifter for said maintenance operations.
18. The computer according to claim 17 further having ECC to parity and parity to ECC converters.
19. The computer according to claim 18 further having error detection and isolation storage, both locally and in a reported operation status location.
20. The computer according to claim 19 wherein the operations are processing in a dual pipe, dual dataflow which allows communication and controls for doublewords with accommodating maintenance address and data spaces.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
October 18, 2007
February 21, 2012
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.